Braces and cleaning

It started with me wanting to clean a shelf. One thing led to the other. Now the entire stock of fretboard blanks is sorted by quality, a couple of new drawers from IKEA mounted on one of the shelves and about 200 bracing blanks are about to be ready. There is still a lot of cleaning left after that, but I will have to postpone it to the future.

When I was cleaning, I came across a banana box with splintered spruce blanks of 100-year-old spruce. They came from an old house where about 60 cm long wooden planks were placed loosely in a panty bottom with wood shavings over as insulation. Most of it was in pine, but there was a lot of spruce as well. All plank pieces were sawn from old forest, some spruce planks have such dense grain that you have to use a magnifying glass to be able to distinguish the direction of the grain! It is noticeable that it is old wood, the spruce is mostly light and a little brittle as old wood becomes. I don't really know if I dare to use it in a top. Some are still strong though. Everyone has a great tap-tone!

My friend Björn Sohlin splintered the wood many years ago, but not everything got the final shape for a brace blank. To take advantage of the splintered spruce pieces, I must first give each piece two flat surfaces at 90 degrees to each other. It was made with a long planer tightened in my vise. When doing this it is important to make sure that you get standing wood in the piece you are planing. Then all the pieces were roughly sawn to 9 mm thickness in my small electrical saw bought from Jula. I think it is awful to use it as it is both dangerous and loud, but better that than hand sawing all the pieces…

The box now contains about 200 brace blanks that still need to be shaped further with a planer and drum sander to my standard 8 x 15 mm. But half the job is done anyway. I filled two garbage bags with shavings and there will be more. What is left is just a fraction of all the wood I started with!

A trio GammelGura

I rarely do more than one GammelGura to one and the same customer at a time, in this batch there were three. I made all three ready as the first in the batch, they are still waiting to be picked up. It is two Levin from 1920 and an old European parlor. The customer actually came with three Levins in numerical order (47232, 47233, 47234), the third with floating bridge was made about a year ago. Very special! If you read in "About Levin" in Melodin 1920 was not a good year for Levin, the premises had burned down in 1918 and they only had 5-10 employees.

Two of the three Levin had floating bridges, the third a fixed bridge. The first one in good condition was allowed to keep the floating bridge, the second with a floating bridge was in a very poor condition and was converted to a permanent bridge in DADGAD tuning. The one with a fixed bridge, also in good condition, got a standard GammelGura conversion. Three identical guitars, but still different.

All three in the batch were intonated. I missed that one of the Levin would have DADGAD tuning when I was doing the measurements and I had to measure the intonation twice. In any case, the mistake was a good experiment. It turned out that the first intonation of the nut for a normal tuning was very similar between the two almost identical guitars with a fixed bridge. This indicates that in a factory you should be able to measure some guitars of the same model, take a mean value and then use that intonation on all guitars of the same model and get pretty good intonation at the nut. In a DADGAD tuning, three strings have a different tuning, the other three have slightly thinner strings in the Newtone Herritage set. The intonation of the nut for the strings with a different tuning was very different (about 2 mm) while the intonation of the thinner strings was almost identical. One can conclude that a measured intoned nut saddle works very well even if you change the string thickness, but not as well if you change the tuning.

I have also got a white board where I will write down everything that is different from a normal GammelGura for the guitars I work with! You learn.

An interesting observation is that the one with a fixed bridge had better quality of the wood, especially in the side and bottom. Those with floating bridges had inferior timber. The tap tone in the top was also better on the one with a fixed bridge, it also sounds best when finished I think. All three guitars were very blonde, but the right color on the bottom and side was much darker from the beginning and hid most of the flaws in the wood. You can see the darker red color as a shadow under the string holder and also a little of the red on the back of the one that was converted to a permanent bridge, otherwise most of the red has faded away.

As usual with Levin parlor guitars, the bottom and top were about half a mm too thick to sound at it's best, about 3,5 mm. They were thinned to just under 3 mm. Both guitars got a new fretboard with a 16 ″ radius and bridge in Madagascar rosewood. I used Stewmac's new Gold frets, which is made of the same metal as the EVO Gold bands, for the first time. Very good frets and better than the Dunlop brass frest I used before I think.

On request from the customer, I adjusted down the string height to 2,3 mm on all of them instead of 2,5 mm on thick E (D) on the 12th band and the usual string height 1,5 mm on thin e (d). This may become a new standard for me, it does not rattle at least when I play.

GG175 with an original floating bridge was in poor condition. It had big cracks in the top that I filled with spruce sticks, the bottom was completely loose and had shrunk unusually much. The top had also been painted on (or rather smeared on!) with some type of plastic lacquer. Fortunately, it was possible to gently scrape loose the new lacquer without scraping through the original lacquer underneath. To be able to glue back the bottom, a 6 mm wide (!) rosewood strip was glued in between the bottom halves. The shade after the string holder is the only problem with converting from floating to a fixed bridge. Just give it another 50 years… Since it's DADGAD tuned I can not play it, but I think it sounds good.

GG174 with an original fixed bridge was in good condition. No cracks except for a small one in the joint between the bottom halves. The neck in birch / maple on both this and GG175 is slightly V-shaped and just below the border to feeling too thick. Nice to play on. The new fretboards were made 6 mm thick, with a 16 ″ radius cut in, they were about 5 mm thick on the edge. I skipped the plate in maple in the bottom of the neck pocket as both guitars came loose nicely in the glue without chipping up the neck block. The neck foot on it was sawn unusually obliquely, but it was compensated by sawing just as obliquely in the neck pocket! Because it was in such good condition, the renovation went without problems.

All guitars in the batch have been given a slightly different bridgeplate in spruce across the entire top. My thought was that the change would provide a faster response. It's hard to say if it turned out that way, the only thing I can say is that these three sound very good and especially this one with a fixed bridge as the original. Maybe the problem with the body's resonant tone decreased.

GG179 was grandmother's guitar, a European made parlor from about 1910. Crack-free except for the classic cracks on both sides of the fretboard glued to the top and with a thin strong V-shaped neck. Unusually, it got plum wood in both the fingerboard and neck. Plum wood is almost as hard as rosewood and as dense as ebony in structure. The heartwood can be as dark as rosewood, but is for the most part a little redder or salmon pink in color. Plum wood is probably the best alternative if you want local woods in a fretboard and bridge. I still think that rosewood is unbeatable when it comes to the sound, this one sounds good but would probably have had a fuller tone with rosewood. As the wood in the fretboard is lighter than usual, I used black side dots.

I had a problem with the top, under the glued piece of the fretboard it was not flat. I had reset the neck one more time when the height of the saddle went wrong, even though I measured carefully before gluing. After putting a spruce shim in the top under the fretboard and sanding the surface flat, it turned out right. In the future, I will keep my eyes open and double check and maybe flatten the top under the fretboard before measuring the angle. A small, small error in the neck angle gives a great effect of the height of the saddle. I have changed from hot hide glue to fish glue and Old brown glue for gluing the neck in this batch. If something goes wrong with the neck set, it is easier to loosen the neck again, especially since they do not dry as quickly as hot hide glue. After 12 hours they are still not completely cured.

May well take some picture when they are handed over eventually 🙂

Proxxon KG 50

To make my segmented saddles, I need four bone posts and two larger bones for the outermost E / e strings.

When bone postst are to be made, I thin the thickness of the saddle blanks to 5,0 mm and a number with a few tenths more or less thickness in my drum sander. The precision when I saw in the spruce piece for the segmented saddle is not completely perfect, it is good to have some alternatives. Then it remains to cut the thinned saddle bone blanks into pieces that are partly 5,5 mm and partly 13 mm wide. I have previously used jigs and a hacksaw for cutting, but it was hardly perpendicular cuts. It was also both laborious and difficult. Something must be done.

I found a small handy circle saw made of Proxxon, and KG 50. It seemed to be perfect for the purpose.

When I got it home, I had to work a little extra to modify it before I could start. There was a 5 mm gap between the blade and the clamp to be able to cut at 45 degrees. I needed to cut the whole saddle bone blank into small 5,5 mm long pieces, so I made some abutments to extend the jaws in the clamp. I used aluminum. In addition, the stop that came with it was very small and swaying and the width 5,5 was at least + -1 mm when I tested. It had to be a real screw as a stop instead.

A problem with the stop was that the cut piece easily got stuck between the stop and the blade. Since the blade has the consistency of hard bread, some broke. Now I use an approximately 3 mm thick aluminum plate between the blank and the stop screw which is removed during cutting. Then a gap is left when the bone post is cut to fall into and it does not get stuck and destroy the blade.

It works very well! Perfect 90 degree cut with the right dimensions and no sweat in the forehead. Each cut takes about 10 seconds. My point extraction (and face mask) is necessary as the sanding pad produces a lot of fine-grained bone dust. The flour is so fine-grained that it cannot be vacuumed off, you must use a damp cloth.


Just got home 200 new 90 x 12,3 x 5,3 mm saddle bone blanks made from camel bone from New Delhi in India. Clearly exotic! This is the third order from no problem. My contact person is Rashdi Malik.



I bought a lot of well used old files at a local auction via a picture online. When I picked them up, they were twice as big as I thought, probably 10 kg of scrap iron! : ) But there were several nice finds among them after all, most were Öberg's quality files. After hydrochloric acid sharpening for 12 hours in 30% hydrochloric acid, they became like new or at least much sharper. A smaller file was perfect for the least enjoyable part of the renovations, to file the ends of new frets. It cuts like butter! The old dull file I used before is now retired.

I geeked into files and found one online book from 1961, “A few words about Öberg's files”. In the middle of the book, there was some good advices on how to use and store files. Having files and rasps together in a box that I had before was not good, so I made a holder for the most used and freshly sharpened files and rasps where no one scratches against each other. With holes and plugs, you can divide the compartments in the holder so they fit different file widths. If I were to redo it, I would make at least three rows of holes with a 3 mm offset between the holes for the best adjustment of the compartments. The longest metal file in the holder was one of the auction files. Now I just need a real file brush!

New date for my rosewood Levin

When I renovated my fine Levin parlor with rosewood at the bottom and sides I was very unsure how old it was. It has no serial numbers stamped. I hesitated between a very early Levin or a manufacture around 1920.

The riddle is approaching a solution when pictures of a Levin lute guitar marked MAY 9 1901, the oldest dating I have seen on a Levin apart from the mandolin guitar from the same year with number 244 that I renovated, just appeared.

This lute has exactly the same ink stamp as my parlor and the same pyramid bridge. It seems that Levin did not get started with his serial numbers on many of the first manufactured instruments. Some got number stamps like the guitar mandolin, others just ink stamps like my parlor and sometimes also as in lute an ink stamped date dating. It is also interesting that MAJ is spelled in English "MAY", Herman got the stamps from the USA.

I did a double check on the original braces in the bottom which were replaced during the renovation of my parlor, but there was no hint of markings. My parlor is very early, perhaps one of the very first to be made. In any case, hardly later than about 1902 and not 1920 as I thought before.

Ongoing batch

It's on it's way. All six guitars in the batch have been repaired, the bottom and ltop have been thinned where needed, the necks have been given a carbon fiber rod, the top and bottom have been given new braces.

Three in the batch have the same buyer, I intend to complete the three in parallel first and maybe do a few small things with the others in the meantime. The focus is on the three that have now had the bottom glued, bridges and fretboards made and each one a K&K pickup. I have also fitted plugs and reinforcements around the stringpin holes.

This is the situation right now. I'm looking forward to hearing if the extended bridge plate gives a quicker response, judging by the Anderberg with a similar bridge plate, there is a good chance that it will be so :-)

Two of three Levins in the same number order from 1920 and a "grandma" guitar, a European curved guitar from about 1910. The Levins get a new rosewood board, the "grandma" guitar gets a mustache bridge and a fingerboard in plum wood.

The other three are a good distance along the way too.

P. A. Anderberg around 1905 with the world's first trussrod

It is not often you get to work with a historical instrument. This parlor guitar made by Pehr A. Anderberg in the USA is one such. It has a fully functional trussrod that was patented by Anderberg as early as 1894 before Gibson patented its variant in 1923. Pehr passed away in 1910 and the patent was invalid when Gibson received his patent. Here is Anderberg's patent from 1894. Gibsons patent from 1923.

In the patent, in addition to the trussrod, there was, among other things, a tricky bridge that actually resembles the one I invented for my very first "renovation" (which I am now a little ashamed of)! Several flies in one go. The "trussrod" in the patent is actually not a rod but a wire.

It all started when my friend Farre (who has been an apprentice in the room twice) sent a link to a guitar for sale on a Swedish auction site, Ekenbergs Auktioner in Karlshamn in Blekinge. The auctioneer had taken very nice pictures and the guitar was very interesting, I wrote "Buy it!" in the reply email.

  • Total length: 95,5 cm
  • Top (upper round, waist, lower round): 23,5 - 19 - 33,5 cm
  • Side (neck block, waist, end block): 9 - 9,8 - 10,8 cm
  • Neck: Soft V-shape
  • Fingerboard (nut, 12th, bridge): 47 - 57 - 59 mm
  • String length: 62,5 cm
  • Lacquer: Spirit lacquer/Shellac
  • Weight: 1274 g
  • Thickness top: 2,8 mm
  • Thickness of the side: 2 mm
  • Thickness bottom 2,5 mm

After a bit of googling on Anderberg, I found this information in an old magazine, The Music Trade Review, from 1904 and also the patent.



[Special to The Review.]

Boston, Mass., Jan 12, 1904. PA Anderberg-, who formerly made guitars and mandolins for John C. Haynes & Co., of this city, now merged in the Oliver Ditson Co., has formed a co-partner-ship with his son, Ralph H. Anderberg until recently a mandolin and guitar manufacturer of Mt. Vernon, NY, for the purpose of making guitars and mandolins. At present the factory of the two Anderbergs (father and son) is located at their home on Cottage street, Chelsea, Mass., But they contemplate moving their workrooms to Boston, in the near future.


By a lucky coincidence, I received an email from Bruce Cowan in the US asking for details on the Bauer guitar from the 1890s that Farre and I renovated during one of his visits to my shop. I quotes some of his emails here.


“George Bauer was a manufacturer, and I do not think he was a luthier. He came to Philadelphia around 1890 as a young salesman for the Boston company John C. Haynes, manufacturers of Bay State instruments and many others.

He started marketing his own make of guitars and mandolins in 1894. His luthier was Ernest Anderberg. In 1898, Bauer partnered with SS Stewart the banjo manufacturer, and then Stewart died. In the catalog linked on your site it shows two lines of instruments, the Monogram and more expensive George Bauer line.

He supplied a Monogram equivalent to the Sears Roebuck catalog company through 1903 or so. They were labeled Acme Professional, and I see one on your page. In 1898, Ernest left the company and his father Per Anderberg came to help ramp up production, as he had done with Bay State guitars in Boston. Per (usually spelled Pehr) was born near Malmo and learned the trade there. I've been in touch with Kenneth Sparr about that… perhaps you know him, as he is a Swedish guitar scholar.

(There were other Swedes at Bay State who went on to form Vega guitars and became VERY successful. Earlier, Per's brother Erland Anderberg started a factory in Mount Vernon New York. By 1900 it had 50 employees. I have not yet found out what labels went on them, but there must have been thousands! It continued until 1915. There's probably another article in these Swedish makers from the Boston area.)

Bauer had a lot of business trouble in 1901 or so, when he cut out the Stewarts, and the company reformed again in 1904 with his brother Emil at the helm. I think the company was gone by about 1911 or so. Emil was involved in the Keen-o-phone phonograph company, and it was gone by about 1914.

I found that Bauer died in 1946 in an insane asylum in Pennsylvania. I found other evidence of mental illness, and it may explain the business trouble in the early 1900s and why he disappeared from the music business. ”


In addition to details about the Bauer guitar, I mentioned the Anderberg guitar in the conversation with him and it turned out that he is probably the one in the world who knows the most about Pehr A. Anderberg! He was one of many in a whole collective of Swedish immigrants to the USA who made guitars of high quality, he was probably one of the most skilled and was hired as a craftsman by several large manufacturers. This is what Bruce writes in an email (where he included the auction guitar, the only one marked P. A. Anderberg that he knows of).


“I think I need to complete the Bauer story before I tackle the Anderberg story. With the Anderbergs I could look at other Swedish makers of that time in Boston, like the Nelsons. 

I have never seen an Anderberg labeled guitar. This is a rare object. Here's the basic PA Anderberg information I've researched. I did not include the footnotes, but I have marked the sources.

Per Andersson Anderberg was born May 21,1838, in Tygelsjo near Malmo, Sweden (Swedish reference). He was an apprentice in the cabinet shop of Carl Johan Bergman in Malmö during the period 1854-1860. (Reference from Kenneth Sparr). He came to America in 1864 (1900 US Census, Ernest's account). He made guitars for C. Bruno, the well-known Manhattan manufacturer and retailer. (Son Ernest's account).

In America, he was known as PA Anderberg or Pehr Anderberg. Records show that Pehr Anderberg married Hulda Elfrida Huppner (b. Sweden, 7 June 1851) in 1872 in Brooklyn. (footnote church record).

The 1873/4 City Directory for New York lists a Peter Anderberg as a guitar maker working at on Pearl Street, in Manhattan, and living in Brooklyn. The directory shows no music retailers or manufacturers listed at that address in that era.

Pehr's brother Erland came to America in 1873 (later passport app), and the two of them made guitars in Mount Vernon, New York and the adjacent settlement of Washingtonville. Ernest Alfrid Anderberg (DATE) and Rolf Hugo Anderberg (DATE) were born at Mount Vernon. Erland manufactured guitars and mandolins there until about 1915.

Pehr and his family moved to Malmo late in the 1870s, where his son Edvin was born in 1880. Pehr had a retail store there (Ernest, Swedish source?) The family returned to the United States in August, 1882 (immigration records).

On their return they lived in the Boston area. PA Anderberg supplied guitars to the John C. Haynes & Co. From about 1888-1891, he was the foreman in their factory.

After that Anderberg went into business for himself in Chelsea, near Boston, manufacturing fretted instruments for August Pollman, a New York retailer. In 1894, he received a U.S. Patent 516717A, an improvement on guitar design, an adjustable stiffening wire that was the forerunner of the truss rod. The only guitar this author has seen with Anderberg's name on it is likely from this period, and it has this improvement. It was sold at auction in Sweden in January 2021.

Around 1898, PA Anderberg began working for Stewart & Bauer in Philadelphia following the merger of banjo manufacturer SS Stewart and George Bauer, a maker of guitars and mandolins. Production was ramped up to supply “Acme Professional” instruments to Sears Roebuck & Co. (Anderberg's son Ernest, who had learned luthiery at his father's side, had been making guitars for Bauer since about 1893.)

In 1904, Anderberg was back in business in Chelsea, either as a repairer, maker, or both. In the summer of 1907, Hulda and Pehr visited Sweden. In April, 1908, Anderberg's shop was destroyed in the Chelsea fire.

Per Andersson Anderberg died in Chelsea on March 18, 1910. By then, his son Ernest had returned to Chelsea, and he continued working there as a repairer for a few more decades.

ATTACHED are a couple of screen shots from the main source. It is not completely accurate. I hope to visit Boston and see the author's papers. Also, the fire is out of historical sequence. I used a lot of maps and other resources.

The main source is viewable here:  Searching the text does not connect with the table in the back of the book.

Anderberg had built "Tilton patent" guitars when he worked for Haynes, and knew that having a popular patent could pay off over the long run. Tilton came up with some interesting innovations, mainly a solid wood rod that was an extension of the neck running through the body to the end pin, taking some stress off the neck joint. These often had a metal disk badge floating in the soundhole, attached to the rod. They also had tailpieces. On a few, the wood grain ran diagonally.

The Anderbergs and Bauer had factories with maybe 30 employees. I see pictures of a three-story brick building. Then the Chicago manufacturing took off with huge plants, and that ended Bay State and Bauer and Erland Anderberg.

Pehr's son did not spend much time in luthiery, as I've seen him in city directories doing other work, but the mention of his work in Mount Vernon seems to indicate that he worked with his uncle Erland at least briefly. It is the only indication I have of a connection between Pehr and Erland after 1880 or so. The son was young at this time (20s) and Pehr was old (66).

Attached is death certificate for 1910.

Pehr's brother Erland Anderberg had a patent too. An adjustable bridge.

Erland had a factory in Mount Vernon, New York from 1882-1915. In 1900 he had dozens of employees. He must have made thousands of instruments, but I can not find out who put their label on them. Probably one of the big New York City retailers. I have not seen any instruments attributed to him ”.


A really fantastically detailed account to say the least, a big thank you to Bruce Cowan!

Pehr and his brother were clearly very knowledgeable and skilled craftsmen and made many of the most beautiful guitars made in the United States around the turn of the century 1900 under different names. This guitar marked P. A. Anderberg was most likely manufactured between 1904 and 1908 before his workshop in Chelsea burned down. Maybe Pehr brought this guitar to Skåne during his visit to Sweden in 1907, the auction in Blekinge is not far from Skåne. Unfortunately, Farre did not get hold of those who sold the guitar, the only thing he learned was that many in the family for the estate were sailors. Probably the history had cooled down and the guitar hung as a wall decoration with its previous owner.

I promised Farre to repair it if he bought it, luckily he got it for SEK 8500. There was a bidding war with another buyer, otherwise it would have gone for less money. It will fit well next to his equally fine Bauer guitar which was perhaps built by Pehr's brother! Well. After a number of weeks, I got the guitar home in a Gator 3/4 case in which it was just possible to fit the guitar. A size comparison with Farres Bauer guitar and a Levin parlor. The shape was very similar to the Bauer guitar.

The guitar was in good condition but not without problems. The bridge was loose and the bottom had a crack in the middle joint and the bottom was loose from the side in some places. The top was deformed around the bridge and the bottom of oak was sunk. The tuners were very strange and as it turned out later unusable when the cog slipped when trying to tune. The neck was bent, the reason was that the trussrod had not been tightened, and it had probably been kept unplayed with tuned up strings for many years. The neck had been re-glued with carpentry glue, there was also carpentry glue in the middle joint of the bottom and at the edge of the bottom. Two brass nails had been nailed into the neck foot… The end block later turned out to be cracked in the middle. The side had received some almost invisible stab wounds in two places, they could easily be glued together with hot skin glue.

I was very curious about the construction of the trussrod, as the neck had to be reset the step was not far to loosen the fretboard to really be able to study the trussrod and make it easier to loosen the neck. It turned out that the neck had a "Levin attachment" without dovetail, it had probably gone well to loosen the neck without taking off the fretboard first.

The trussrod was not a wire as in the patent, instead it looked pretty much like a normal Gibson type trussrod. I guess it did not take long before Pehr realized that a rod works much better than a wire! The trussrod has its problems as it protrudes into the fretboard at both ends. The attachment at the neck foot is strong, the bar has been bent 90 degrees, which can be seen in the pictures, but the nut can not be tightened too hard as it only rests on the ebony nut and the end wood on the fingerboard. On this particular guitar, the trussrod does not work so well as the wood of the neck is both strong and hard. The most important thing is that it is tightened and not loose, you can not bend back the neck relief more than a little bit. I specially made a fixed 8 mm wrench to access the nut which was very tight. A crescent-shaped hole for the trussrod was recessed on the underside of the nut.

It was easy to loosen the bottom. The braces at the bottom were relatively loose, the glued spruce to oak does not sit as well as spruce to spruce. However, the braces in the top were really tight! I like the bracing in the top, it is similar to the one I use in my GammelGura. I actually got some inspiration and nowadays, I stick the ends of the braces under the kerfing on GammelGura guitars and have modified the bridge plate so it goes over the entire top.

It is noticeable that it was a skilled craftsman who built the guitar. An extremely beautiful bridge and end pin, nicely and sensibly put together otherwise.

As I said, the top was deformed around the bridge and also the bottom, I tried to get it in some pictures. The bottom still curved more when it was loosened from the side, the oak had shrunk and the glue in the bottom braces held up.

To flatten the top and bottom, the braces were removed from the bottom and around the bridge in the top. The top and bottom were thoroughly soaked with water and put under pressure for a few days.

The result was good.
The original braces at the bottom were deformed and bent in the wrong direction, I replaced one that I did not get loose whole with a replica with the same dimensions, but with a radius in the other direction. The other original braces were shaped on the underside to a radius in the right direction. An extra center stick was fitted to hold the center joint in the bottom together.
Two of the braces in the top cracked along the grain when I tried to loosen them due to run-out. They were replaced with new replicas of the same size. I could not help myself and replaced the bridge plate and the two reinforcements in thin maple with spruce. Round maple reinforcements were glued around the stringpin holes. I also made the bridge plate a little wider for a possible future installation of a K&K pickup. With the new braces, the top had a slight bend upwards, especially the new brace under the fingerboard was good as the neck angle to the bridge got better.

The bottom could be glued without major problems, part of the edge of the bottom had to be scrapped away in the narrowest place as usual. The shrunken bottom has a smaller circumference than the sides, a simple solution is to push the sides in a few mm in the narrowest place for the best fit.

The fretboard was glued back without any problems.

Because the trussrod stuck up from the neck at front and back, I could not sand the neck straight. Instead, I tried to heat and bend back the neck with a loose trussrod and about 90 degrees C of heat (as hot as I dared to use not to melt the varnish on the back of the neck). It got better, but not quite straight. The neck, made of some kind of fruit tree or possibly maple, was stiff and hard and did not want to bend back.

The neck was glued on. Despite all the measurement, it was not perfectly glued in, the biggest reason was that the fretboard was not completely straight. When the frets were removed, I sanded the board that had a 10 ″ radius plane. Unfortunately I had to sand at the top of the fretboard but also at the 12th fret. When the fretboard was straight, the string height of the 12th fret became half a mm too high. I reglued the neck a second time and got to the angle perfectly. A fret at the top of the bridge instead of a saddle places high demands, there is virtually nothing to correct if the neck angle is not perfect.

The bridge, still with its original pins, was re-glued with hot hide glue. The bridge's stringpin holes had deep worn notches from the strings, they were also not centered in the hole. I used tape as "a dam" for superglue and sanding dust of rosewood that filled the grooves. It worked well!


The fretboard, which was sanded flat, was re-fretted with new and higher nickel frets. The originals were saved in the case, but they were soft and not of the best quality. The light colored parts of the fretboard are the areas that were not sanded down.

To intonate better, the fret groove was filled in the bridge and a new fret was mounted 2 mm further down on the bass side and 1 mm further down on the treble side. Some previous damage on the top around the bridge was hidden with golden brown stain, the scrapped edge of the oak bottom was repaired with spirit varnish and a thin brush. New Golden Age tuners with black knobs were fitted.

All replaced parts are saved in the case for the guitar. After four days of vibration, it sounds open and loud with low tension Newtone Heritage 0.11 strings.The intonation on the 12th fret is good. Compared to an GammelGura with all the features, the tone is more brutal and more primitive than I am used to. The metal fret "saddle" on the bridge and the nut in ebony are not optional for the tone. The body in oak, on the other hand, is more resonant than expected. 100 years of drying will probably do its thing!

Improved bracing in the top

I have been thinking a bit about how I brace up the top in GammelGura guitars. It was time to improve my methods.

As I recently made a go-bar jig, new possibilities have opened up. Among other things, I can glue braces and spruce tiles to a surface that is bent in both directions, both across and along the guitar. With my old cauls, I got the right bend only across the guitar. Which is optimal for the large braces, but not the tiles with a larger adhesive surface. You can also glue several complex pieces at once as the go-bar has bars that take up less space.

I am renovating a rare PA Anderberg parlor from around 1905 and the USA. Some original braces were glued to it and I noticed that it was actually easy to put the ends of the brace in recessed holes under the kerfing. A big advantage is also that the brace is attached without glue while measuring the location of all the details of the top. Above all, it was nicer than my previous stop blocks on the ends of the braces! The guitar also had a thin bridgeplate in maple across the top that tapered on either side of the bridge, like the one in a Levin parlor but done better.

Another detail that I have considered is the mandatory cracks in the top on one or both sides of the fretboard. The edges of the fretboard almost always end up along an annual ring in the wood of the top and become an point where a crack can occur. When the wood of the fingerboard shrinks more than the wood of the top (especially if it is ebony) and / or when the neck begins to rotate towards the sound hole from the tension of the string, you get cracks in the top at the edges of the fingerboard.

The six GammelGura in the ongoing batch have received a new variant of the bracing pattern. Here is a picture of one of them with unglued and untrimmed braces and plates.

I numbered the different parts with a pencil. The yellow tape on the spruce plate at the top is to protect the spruce plate from smudges of hot hide glue.

All ends of the transverse braces have been inserted under the kerfing. It is not difficult to cut small gaps with a knife, chisel and scrape out the wood with a curved scraper knife that is otherwise used to clear saw tracks for frets in a fretboard. It is actually less work than gluing everyone on the small stop blocks I used before! The ends of the bridge plate wings, on the other hand, are not tucked under the kerfing but are thinned down to nothing at the end so as not to tie up the top unnecessarily

All braces and plates are made of spruce with standing grain. The underside of all braces and spruce plates is given a small 30 ″ radius on the underside, the same radius as the bowl in the counter in the go-bar.

Spruce plate 1 (with the yellow tape) is 1,5 mm thick and has the grain across the grain of the top. As the oblique edge passes several of the top's annual rings and also the edges of the fretboard, there will be no stress point for a crack in the top. As the fingerboard shrinks, the tension is distributed over several of the grains in the top, hopefully it will not result in the obligatory cracks with time. In order not to pre-tension the top when gluing the fretboard, I also use Old Brown Glue (hot hide glue with added urea) instead of hot hide glue as before between the fretboard and top. OBG does not shrink when it dries as hot hide glue and fish glue do.

The A-frame braces 2 and 3 are rectangular and 6 x 6,5 mm. Both have been jacked out 1,5 mm on top of spruce plate 1, which means that they are 5 mm thick above brace 4. Spruce plate 1 must be intact to do its job in the best way.

Brace 4 is rectangular and is shaped from a blank that is 8 x 15 mm. The ends are thinned down from 6 cm from the end to about 1 mm thick at the far end. I make two gaps on the underside of the brace that exactly fit the A-frame braces 2 and 3

Brace 5 is triangulated from a blank that is 8 x 15 mm. The ends are thinned down from 7 cm from the ends and small outlets are made to insert the ends of the A-frame ribs.

The unnumbered spruce plate between braces 5 and 6 is 1,5 mm and thinned to nothing at the edges. The edges of this plate also pass several grains in the lid so as not to give a stress point for a crack.

Brace 6 is triangulated from a blank that is 8 x 15 mm. The ends are thinned down from 7 cm from the ends.

The spruce plates 7-9 for the cross shaped bridge plate are all in 3 mm thick spruce. The wings 7 and 9 have the grain across the grain of the top, while the central plate, about 8 x 6 cm, under the bridge has the grain along the top for best rigidity. The wings are 3 cm wide up to the edge of the bridge where it narrows to 1 cm. My hope is that the guitar top responds faster when playing as vibrations travel about 4 times faster along the grain than across. The wings also hold the top together and reduce the risk of cracks in the edge of the bridge ends, especially if it is a pyramid bridge with a straight edge.

Brace 10 is 3 mm thick and about 0,5 x 20 cm, the length varies depending on the top shape. On some European parlor guitars it is not needed as the bridge is close to the end block.

The bridge plate with its wings is glued together separately with hort hide glue, the joints are ground at 45 degrees to make them stronger. After about 20 minutes, it can be roughly shaped, the underside sanded to a radius and the edges chamfered. The central plate is given a shape similar to the profile of an aircraft wing.

All braces and plates can be glued in a single session in the go-bar jig. I use abutments so as not to crush the soft spruce and distribute the pressure from the go-bar ribs, they are placed in the right order before I start gluing. The abutments for larger surfaces have a soft 4 mm yoga mat in rubber on the underside to distribute the pressure over the entire surface. Special aluminum brackets are made for the triangular ribs and the sloping ends of the ribs. The large abutment on the aircraft wing-shaped central part of the bridge plate has been traced to be extra flexible. The flat and thin brace 10 has only a strip of transparent plastic on top.

The A-frame braces are guided in from the side through the holes in the brace 4th. There will be a lot of smudges with the hot hide glue, which is why I protect the spruce in plate 1 with low-tack tape. I always use a thin damp cloth, a table knife and a thin palette knife to wipe off excess glue.

It will be a whole forest of go-bar sticks when you are done! My bars are 8 mm birch sticks that are flattened down to about 6 mm with the help of my drum sander. Each part that is glued is heated with a heat gun, set at about 300 degrees C, to extend the gluing time. Previous glue can also be reheated, one of the many benefits of hot hide glue.

It remains to be seen whether the extended wings on the bridge plate gives the desired result, if nothing else, the extensions helps to hold the fibers of the top together.

Braces shaped and sanded.

Suction table

A very important detail for the best sound in an GammelGura is to make sure that both the top and the bottom have the right thickness. Levin parlor guitars in particular usually have too thick tops and bottoms. The top can sometimes be 4 mm thick, the bottom is typically 3,5 mm but can be 4 mm as well. European and American parlor guitars almost always have the right thickness, but it happens that some European guitars have a 3,5 mm thick bottom and American for a thin top of 2.0 mm.

With each new batch, I thin out thick tops with small planers and a sanding mouse. Thick bottoms I thin with my drum sander. I recently bought a larger and wider drum sander that I used for the first time last batch. It was about to end in disaster when some of the bottoms got stuck… luckily it was possible to repair the mistakes with inlays of 0.6 mm maple veneer. It also turned out that the drum sander grinds more on the ends of the bottom when I feed it in. There are two steel rollers on either side of the large roller with sandpaper to press the object down against the conveyor belt. When the bottom is only pressed down by one roller, the bottom springs up and the sander removes a few tenths mm more, which makes the bottom a bit too thin at the ends.

That was a problem. I thought about it and the solution was to make a suction table. With a vacuum from my industrial vacuum cleaner, you should get enough power to keep the bottom flat and safe in the drum sander. Instead of buying an expensive suction table, I made one myself, as I did not know if it would work in practice.

A strong wooden plate was made, the ducts inside were made of strips of wood that were glued in at regular intervals and which were then milled the other way to get a grid of air ducts.

Powerful vacuum hoses and handy nipples were available for purchase in a shop in town, the hoses can be easily disconnected by pressing the orange hat on the nipple. To make the flow greater for the vacuum cleaner and distribute the suction, I mounted 5 hoses, I do not know if it was necessary.

An aluminum plate had a whole bunch of small 1,5 mm holes drilled, the holes end up in the middle of each intersection between four squares. Since only guitar bottoms are to be run in the drum sander this way, the pattern was made in the shape of a guitar body. It was a bit of work to drill all the holes! I also made a rubber mat with punched holes. A feature is that the vacuum around each small hole is evenly distributed as a suction cup in the small volume created by the rubber mat.

To connect the vacuum cleaner to the five air hoses, I bought a plumbing connection with two lids at a local hardware shop. A large hole was drilled in one lid for the vacuum cleaner hose and five more nipples was mounted in the other lid.

This suction table was already made during the Christmas weekend, but only came into use now in the ongoing batch. Two Levin and two European guitars had too thick a bottom. They were all about 3,5 mm thick, the ideal measurement is about 2,5-2,7 mm for a maple bottom. Then you get the best sound but basically the same strength as a thicker bottom.

I had bought some cardboard which I cut as a mat around one of the bottoms. The idea was to cover all the holes that ended up outside the bottom. It was a failure when the air leaked in between the bottom and the cardboard. The solution was to use tape to cover all the holes outside the bottom.

The next problem was that the bottom was not flat but bowl-shaped by the braces glued in with a radius. I got the bottom to attach by pressing hard all around, but the vacuum suction could not hold the bottom in place. I felt like it would not work. BUT. By really soaking the bottom, the wood swelled and the bottom became both more flexible and flatter. It sat like a slap in the vacuum table and it was impossible to loosen it without turning off the vacuum cleaner! When it attaches, the sound of the vacuum cleaner changes and it has to fight harder. A rough vacuum cleaner is not cooled by the air that is sucked in but by a separate fan, so I think it should withstand 10 minutes of hard driving without breaking. Remains to be be seen.

Then it was just to test it all. I first fine-tuned the drum sander so that it took evenly over the entire surface. The hoses protrude on the open side in the drum sander, it was very easy to grind the bottom to the thickness I wanted. By turning off the vacuum cleaner and loosening the bottom from time to time, I was able to control the thickness.

The result exceeded expectations. No problems with thin ends and no drama. Almost perfect thickness on the entire bottom. This bottom was thinned almost 1 mm, but you can still see where the braces were glued.

Another bottom caused problems as it was full of hide glue that clogged the sandpaper in the drum sander. But it was possible to loosen the sandpaper and clean it with water and a dish brush.

It turned out that the tape was just like the one on rubber so I switched to the yellow low-tack tape instead of the packing tape. If you have several bottoms, you start with the largest and then supplement with tape for smaller bottoms where new holes end up outside the bottom.

I am very happy with my suction table and will use it in the future.